ES_corlec_DeepESP-gpt2-spanish
This model is a fine-tuned version of DeepESP/gpt2-spanish on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.0360
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- num_epochs: 7
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
4.2471 | 0.4 | 2000 | 4.2111 |
4.1503 | 0.79 | 4000 | 4.1438 |
4.0749 | 1.19 | 6000 | 4.1077 |
4.024 | 1.59 | 8000 | 4.0857 |
3.9855 | 1.98 | 10000 | 4.0707 |
3.9465 | 2.38 | 12000 | 4.0605 |
3.9277 | 2.78 | 14000 | 4.0533 |
3.9159 | 3.17 | 16000 | 4.0482 |
3.8918 | 3.57 | 18000 | 4.0448 |
3.8789 | 3.97 | 20000 | 4.0421 |
3.8589 | 4.36 | 22000 | 4.0402 |
3.8554 | 4.76 | 24000 | 4.0387 |
3.8509 | 5.15 | 26000 | 4.0377 |
3.8389 | 5.55 | 28000 | 4.0370 |
3.8288 | 5.95 | 30000 | 4.0365 |
3.8293 | 6.34 | 32000 | 4.0362 |
3.8202 | 6.74 | 34000 | 4.0360 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.1+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
- Downloads last month
- 25